Coupling Parameter Identification Method And Its Application Research In PMSM

The coupling identification method is a new identification idea in recent years.It is mainly used to study the parameter identification of linear or nonlinear multivariable systems with complex structure and coupling relationship between parameters.The main principle is to reduce the redundant estimation of the parameters of each subsystem and the redundancy calculation of covariance matrix through the coupling relationship of the parameter estimation of the multivariable subsystem,so as to improve the convergence of the identification algorithm and the estimation accuracy.The electrical parameters of permanent magnet synchronous motor are the basis of its current control regulator design.Moreover,it has great engineering value and theoretical significance for the optimization design of motor position controller and speed controller parameters.Under different operating conditions,the PMSM parameters are easily affected by the operation environment.The change of motor parameters will directly affect the vector coordinate transformation,which greatly affects the control performance of PMSM,so it is necessary to study the dynamic identification of PMSM parameters.Therefore,this paper will take the idea of coupling identification as the basis,take PMSM parameter identification as the research object,combine the multiple innovation identification principle,the finite data window identification principle and the deviation compensation identification method.The dynamic multi parameter identification problem of PMSM and its control application based on its parameter identification are studied.The main research contents are as follows:(1)Based on the mathematical model of PMSM in synchronous rotating coordinates,the identifiability of PMSM electrical parameter least squares is proved,and the continuous excitation conditions for the identification of motor parameters are given,a PMSM experimental platform based on d SPACE DS1007 PPC is constructed also.It lays the foundation of the parameter identification and control of permanent magnet synchronous motor for this chapter and its subsequent chapters.(2)Based on the multiple innovation identification theory,combined with the coupling identification idea and forgetting factor technology,a coupled multi innovation oblivious factor least square identification algorithm is proposed.The multi new technology used in the algorithm can maximize the use of the acquired motor identification information to overcome the lack of data in the identification of motor parameters.At the same time,the coupling identification technology can make full use of the process data,and it has the advantage of avoiding the inverse operation of the high dimensional matrix.The forgetting factor added enables the algorithm to dynamically track the change of PMSM parameters.Simulation and experimental results verify the effectiveness of the proposed algorithm.(3)Based on the finite data window identification theory and the coupling identification idea,the coupled finite data window band forgetting factor least squares identification algorithm is proposed to identify the multi input multi output and dynamic time-varying PMSM multi parameters.Compared with the traditional multivariable band forgetting factor limited data window recursive least square algorithm,the algorithm can reduce the redundancy estimation of the parameter vector of the subsystem by the coupling relation of the parameter estimation of the subsystem,and also reduce the calculation amount of the identification algorithm,thus speeding up the convergence speed of the algorithm.Besides,the algorithm retained the advantage of finite data window least squares algorithm combined with forgetting factor to track the time-varying parameters more quickly.Simulation and experimental results verify the effectiveness of the proposed algorithm.(4)Based on the error compensation band forgetting factor least square algorithm,combined with the coupling identification idea,a partial coupling error compensation band forgetting factor least squares identification algorithm and the coupling error compensation band forgetting factor least squares identification algorithm are proposed.The algorithm can obtain accurate estimation values of PMSM parameters without considering the noise model,and the experimental verification of parameter identification based on PMSM's state equation model and multiple pseudo linear model is carried out respectively.(5)Based on the identification of PMSM parameters,an indirect adaptive backstepping control strategy of PMSM is proposed to solve the problem of dynamic timevarying parameters of PMSM under actual control.In this method,the parameters of PMSM are identified by the multiple new band forgetting factor least squares algorithm,and then the PMSM parameters obtained are fed back to the backstep controller to overcome the dynamic changes of the PMSM parameters and improve the control performance of the PMSM.The stability proof of PMSM control system is given based on the Lyapunov stability theory.Simulation and experimental results verify the effectiveness of the proposed algorithm.